Molecular-level Modeling Of Fcc Slurry Oil Conversion

FCC slurry oil is the heaviest liquid product from fluid catalytic cracking units.Because it is rich of heavy aromatics,metal,carbon residue and catalyst particles,slurry oil is very difficult to further upgrade,and usually is burned directly as fuel oil.However,with the feedstock of FCC unit becoming heavier,how to exploit the slurry oil to get extra value worth deep researching.In this thesis,the supercritical fluid extraction coupling with hydrotreating process can effectively improve the slurry oil properties.First,by means of supercritical extraction,the raw material slurry oil is extracted.After extraction of the slurry oil,most of the catalyst particles,metal can be remove.And the CCR,resin,asphaltene content can be reduced apparently.The extracted slurry oil can meet the requirements of the feedstock for the hydrotreating unit.After the extraction of the slurry oil,the extracted oil was used as the feedstock of hydrotreating unit.Then the pilot hydrotreating experiment was carried out,and the molecular-level kinetic model of the hydrogenation process was built.It is a hot researching area that developing kinetics models at molecular level.The first thing of molecular-level kinetic modeling is to develop the molecular storage method of computer,the previous literature proposed a variety of storage methods.But there exists many issues in these methods,including complicated coding and multiple molecular structure can be denoted by one storage format.In this thesis,by introducing cheminformatics method,SMILES language was used to store molecules.This method can solve many problems of literature methods.This thesis discusses two methods,stochastic construction method and information entropy maximization method.And a method was proposed to improve the construction efficiency by predefining the molecular structure library.The entropy maximization algorithm was used to construct the feedstock molecular set of the extracted FCC slurry oil by SFE.The results were consistent with the experimental data.In the process of hydrogenation reaction network construction,the reaction SMARTS language is used to solve the problem of coding of reaction rules.After the reaction network is generated,it needs to be transformed into the form of the ordinary differential equations which can be solved by the computer.Then the processing method of the stoichiometric coefficient matrix is used to convert the reaction network into a computer-solvable form.Finally,by using the model construction method,the molecular level kinetic modeling of the hydrocracking process of the slurry is carried out.The results of the model can be consistent with the experimental results.After the building of the model,there may still exist problems with the resulting model.Or the model needs for further analysis.The network visualization method can be used to analyze the whole construction process and the kinetic model results.Reaction network visualization has been widely used in biological area.By integrating with property of species in reaction network,the variation of this property can be interpreted clearly.According to the feature of chemical reaction network,the species-species and species-reaction network are chosen.Then by integrating the generation and content of species into reaction network,a series of visualization method are developed.The layout of the network is discussed in detail,and a variety of network layout schemes are analyzed.Finally,the tree layout and the force layout are selected as the final layout of the reaction network.Based on the visualization,the visualization and reaction network data files are integrated and can be directly edited the reaction network visualized interface.In this thesis,the molecular-level kinetic model of FCC slurry oil was built.At the same time,a model building platform of molecular-level kinetics was developed,which could be used to construct the molecular level kinetic model quickly.Related software developing library,the basic framework of software and part of the code were briefly introduced.